JP4710417B2 - Information processing device - Google Patents

Description

The present invention relates to an information processing apparatus .

In recent years, it has become difficult for users to acquire desired information while a large amount of information is distributed due to the development of communication devices. Therefore, Patent Document 1 discloses a technique for extracting proper nouns from an electronic document based on the viewpoint of obtaining desired information from specific words. Specifically, a pattern that captures features of proper nouns is stored in advance, whether or not it is a proper noun is determined based on whether or not it matches the pattern, and proper nouns are extracted based on the result. .
JP-A-6-52221

  By the way, with the recent globalization, opportunities to come into contact with documents (electronic documents) other than their native language are increasing. Therefore, even if the document acquired by the user is a desired document, if the document is described in a language other than the user's native language, it is desired if the user has already mastered the language. Although it can be recognized as information, if the user is not proficient, the contents of the document cannot be understood, and the information cannot be recognized as desired. As a result, there is a possibility that information that the user wants to acquire cannot be acquired.

  In the technique of the above-mentioned Patent Document 1, since only proper nouns are extracted, even if documents with the same content written by the same person are written in different languages, each document Proper nouns written in the language are extracted. Therefore, for example, when a user searches the database using “Mr. A”, which is a proper noun, as a key, a document obtained as a search result differs depending on which language “Mr. A” as a key is expressed. Further, when the language of the document including the proper noun stored in the database is different from the language of the proper noun used for the search, the technique disclosed in Patent Document 1 is desirable even if the proper noun is the same person. Documents are not searched.

  The present invention has been made in view of such circumstances, and an object thereof is to provide a technique for extracting a proper noun from a document and searching for a document written in a different language based on the extracted proper noun. .

The present invention relates to first storage means for storing a Chinese name of a person name and an English notation of the person name in association with each other, and a first extraction means for extracting a person name from image data representing a manuscript written in Chinese. Conversion means for converting a person name extracted by the first extraction means into a person name written in English based on the stored contents of the first storage means; and accepting means for receiving input of a file name of the manuscript A second storage unit for storing the file name of the document received by the reception unit, the person name extracted by the first extraction unit, and the person name converted by the conversion unit in association with each other; Second extraction means for extracting a person's name from the image data representing the written original; a file name of the original associated with the person name extracted by the second extraction means; and the first extraction means. An information processing apparatus comprising: search means for searching for the extracted person name from the second storage means; and display means for displaying the file name and person name searched by the search means. provide.

  FIG. 1 is a block diagram showing a configuration example of a document search system 1 including a search device 20 which is an embodiment of a document search device according to the present invention. An image reading apparatus 10 in FIG. 1 is a scanner apparatus having an automatic paper feed mechanism such as an ADF (Auto Document Feeder), for example, and optically reads a paper document set in the ADF page by page, and converts the read image into an image. Corresponding image data is transferred to the search device 20 via the communication line 12. As the communication line 12, a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, or the like can be used. In this embodiment, a LAN is used as the communication line 12.

FIG. 2 is a block diagram illustrating a hardware configuration of the search device 20.
The control unit 21 is, for example, a CPU (Central Processing Unit), and centrally controls each unit of the search device 20 by executing various kinds of software stored in the storage unit 25. The communication IF unit 22 is connected to the image reading apparatus 10 via the communication line 12, receives image data transmitted from the image reading apparatus 10 via the communication line 12, and transfers the image data to the control unit 21.

  The display unit 23 is, for example, a liquid crystal display and its drive circuit, and displays an image corresponding to the data transferred from the control unit 21. The operation unit 24 is, for example, a keyboard provided with a plurality of operation elements (not shown), and outputs data (hereinafter, operation content data) corresponding to the operation contents of these operation elements to the control unit 21.

  The storage unit 25 includes a volatile storage unit 25a and a nonvolatile storage unit 25b. The volatile storage unit 25 a is, for example, a RAM (Random Access Memory), and is used as a work area of the control unit 21. The nonvolatile storage unit 25b is, for example, a hard disk. The non-volatile storage unit 25b stores the proper noun dictionary shown in FIG. As shown in FIG. 3, the proper noun dictionary is composed of a “Chinese notation” field and an “English notation” field, and the Chinese notation for the proper noun and the English notation are registered in association with each other. For example, since the English notation of “Tayoshi” is “Tian Fang”, “Tian Fang” is stored in association with “Tayoshi”, and the English notation of “Hakuro” is “Bai Lang”. “Bai Lang” is stored in association with “Hakuro”.

  The nonvolatile storage unit 25b stores the file creator table shown in FIG. 4, OS software for causing the control unit 21 to execute an OS (Operating System), and proper noun search software. As shown in FIG. 4, the file creator table includes a “file name” field and a “file creator” field. The “file creator” field further includes a “Chinese notation” field and an “English notation” field. It is constituted by.

  Next, operations performed by the control unit 21 will be described with reference to FIG. When the power source (not shown) of the search device 20 is turned on, the control unit 21 first reads the OS software from the nonvolatile storage unit 25b. When the execution of the proper noun search software is instructed by the user's instruction, the control unit 21 reads the proper noun search software from the nonvolatile storage unit 25b and executes it. Here, there are mainly two operations of the control unit 21. The first is an “image data storage” operation, and the second is an “image data search” operation. The image data accumulation operation is an operation for accumulating the image data transferred from the image reading device 10 in the nonvolatile storage unit 25b. The image data search operation is an operation for searching the image data stored in the nonvolatile storage unit 25b described above based on specific information. These operations are selected by the control unit 21 in accordance with a user instruction.

  First, the control unit 21 displays the selection menu 30 shown in FIG. 6 on the display unit 23, and allows the user to select the image data processing method (step SA10). As shown in FIG. 6, the selection menu 30 includes an “image data storage” button BT31 and an “image data search” button BT32. When the “image data storage” button BT31 is selected and pressed by the user, the control unit 21 receives operation content data representing the user's operation content from the operation unit 24, and specifies a processing method based on the operation content data. . Here, by receiving the operation content data indicating “image data accumulation”, it is specified that the selected processing method is “image data accumulation”. Then, the control unit 21 waits for image data to be sent from the image reading apparatus 10.

  On the other hand, when a user sets a paper document in the ADF of the image reading apparatus 10 and performs a predetermined operation, an image representing the description content of the paper document is read by the image reading apparatus 10 and image data corresponding to the image is communicated. The data is sent from the image reading device 10 to the search device 20 via the line 12.

  When the control unit 21 receives the image data sent from the image reading device 10 via the communication IF unit 22 (step SA12), the control unit 21 stores the received image data in the nonvolatile storage unit 25b and analyzes the layout of the image data. To recognize the character area (step SA14). Then, each recognized character area is subjected to OCR processing to form text data (step SA16). Next, the control part 21 extracts a proper noun from the text data produced | generated in step SA16 using a morphological analysis and a proper noun extraction technique. Proper noun extraction technology assigns tags representing person names, organization names, and place names to each person name, organization name, place name, etc. in text data that has been subjected to morphological analysis, and extracts a phrase with the specified tag. Technology. In the present embodiment, in particular, a person name is extracted by this proper noun extraction technique, and a person name matching the proper noun stored in the proper noun dictionary shown in FIG. 3 is extracted from the extracted person name (step SA18). .

  Here, step SA18 will be described in more detail. When the image data sent from the image reading apparatus 10 is the document 50 shown in FIG. Is extracted. In FIG. 7, the part indicated by “OO” indicates that the document is written in Chinese.

  Next, since the image data processing method designated by the user is “image data storage” (step SA20; YES), the control unit 21 selects the proper noun “Tayoshi” in Chinese notation extracted in step SA18. Then, using the proper noun dictionary shown in FIG. 3, the proper noun “Tian Fang” written in English is converted (step SA22). Then, the file name input screen 40 shown in FIG. 8 is displayed on the display unit 23, and the user inputs a file name by the operation unit 24 (step SA24). As shown in FIG. 8, the file name input screen 40 is provided with an input box 41, and the user inputs a file name in the input box 41 by operating the operation unit 24.

  When the control unit 21 receives the operation content data including the file name from the operation unit 24, the file name, the proper noun in Chinese notation extracted in Step SA18, and the proper noun in Chinese notation are expressed in English in Step SA22. The proper nouns converted into are stored in the file creator table shown in FIG. 4 (step SA26). Specifically, if the file name input by the user in step SA 24 is “FileX”, “FileX” is stored in the “filename” field of the file creator table. Then, “Tayoshi” and “Tian Fang” are stored in the “Chinese notation” field and the “English notation” field of the “file creator” field, respectively. By doing so, the file name and the Chinese notation and English notation of the creator who created the file are stored in association with each other.

  Next, the operation of the control unit 21 when the user presses the “image data search” button BT32 of the selection menu 30 (see FIG. 6) displayed on the display unit 23 by the control unit 21 in step SA10 will be described. At this time, it is assumed that a sufficient amount of data is stored in the file creator table (see FIG. 4).

  The control unit 21 receives the operation content data indicating “image data search” from the operation unit 24, thereby specifying that the selected processing method is “image data search”. Then, the control unit 21 writes and stores in the volatile storage unit 25 a that the specified processing method is “image data search”, and waits for image data to be sent from the image reading device 10. At this time, it is assumed that the paper document set in the ADF of the image reading apparatus 10 is written in English. When a paper document is set by the user, as described above, an image representing the description content of the paper document is read by the image reading device 10, and image data corresponding to the image is read via the communication line 12. To the search device 20.

  The control unit 21 performs the above-described step SA12 → step SA14 → step SA16 → step SA18, and extracts proper nouns from the text data.

  Here, step SA18 will be described in detail. When the image data transferred from the image reading apparatus 10 is the document 52 shown in FIG. 9, the control unit 21 reads a document in which the transferred image data is written in English. Recognize that the image data represents. Then, layout analysis and OCR processing are performed on the image data, and “Tian Fang” is extracted as a proper noun from the text data by a proper noun extraction process. In FIG. 9, the part indicated by “ΔΔ” indicates that it is written in English.

  Next, in step SA10, the processing method of the image data designated by the user via the selection menu 30 is “image data search” (step SA20; NO). Therefore, the control unit 21 extracts in step SA18. Based on the proper noun “Tian Fang”, the “English notation” field of the “file creator” field in the file creator table shown in FIG. 4 is searched (step SA28). If there is a corresponding proper noun (step SA30; YES), the file name stored in association with the proper noun and the Chinese representation of the proper noun are displayed on the display unit 23 (step SA32). In this case, “FileX” (“file name” field) and “Tayoshi” (“Chinese notation” field in the “file creator” field) are stored in association with “Tian Fang”. The control unit 21 displays “FileX” and “Tayoshi” on the display unit 23 as search results. As a result, the user can recognize that “Tian Fang” in English and “Tayoshi” in Chinese are the same person, and files created by the person have already been accumulated. Can be grasped.

  On the other hand, when the file creator table is searched based on the proper nouns extracted from the image data in step SA28, if there is no corresponding proper noun (step SA30; NO), that fact is displayed on the display unit 23 (step SA30). Step SA34).

  As described above, according to this embodiment, when storing a file, a proper noun is extracted from the image data representing the file, and the file name of the image data and the Chinese and English notation of the file creator Are stored in association. Further, when searching for a file, a file stored based on proper nouns extracted from image data representing the file is searched. Therefore, when a user who is not proficient in a language other than their native language obtains a document written in Chinese or English written by the same person, the document created by the creator is searched using the creator as a key. can do.

  Further, when documents having the same content written in Chinese and English are applied to the present embodiment, the user determines that the creators of the respective documents are the same from the search results searched using the respective creators as keys. It becomes possible to estimate. For example, even if the contents of the document in the area 50a indicated by “XX” in FIG. 7 and the contents of the document in the area 52a indicated by “ΔΔ” in FIG. 9 are the same, they are described in different languages. If so, the user may misjudge that these are documents with different contents. However, if it is possible to grasp that the creators of these documents are the same using the creator as a key, the user can assume the relevance of these documents.

  In the above-described embodiment, the document when the control unit 21 performs “image data storage” and the document when the “image data search” are performed are expressed in English as the first language and the second document, respectively. The case where the language is written in Chinese is shown. However, the pattern is not limited to this pattern, and the first language and the second language may be written in Chinese and English, respectively.

  In the present embodiment, the first language and the second language are English and Chinese, respectively. However, the present invention is not limited to English and Chinese, but may be other languages such as French and Italian. May be. At this time, a proper noun dictionary showing the correspondence of proper nouns in each language is held in the nonvolatile storage unit 25b.

  Further, the language associated by the proper noun dictionary is not limited to two languages, and languages of three languages or more may be associated. At this time, as shown in FIG. 10, a proper noun dictionary in which proper nouns expressed in languages of three or more countries are associated is stored in the nonvolatile storage unit 25b. Then, the control unit 21 causes the display unit 23 to display the language selection menu 60 shown in FIG. 11 as language selection means for selecting the first language and the second language. The user who visually recognizes the language selection menu 60 can select a language by operating the pull-down menu 61 using the operation unit 24, and can select a desired language by pressing the enter button BT62. When receiving the operation content data including the selected language from the operation unit 24, the control unit 21 specifies the first language and the second language from the operation content data. Then, from the text data representing the document written in the first language or the second language, the proper noun corresponding to the first language or the second language is extracted based on the proper noun dictionary shown in FIG. To do.

  In the present embodiment, the case where the image reading device 10 and the search device 20 are configured as separate hardware has been described, but both may be configured as integrated hardware. At this time, the communication line 12 serves as an internal bus that connects the image reading device 10 and the search device 20 within the hardware.

  In the present embodiment, layout processing is performed on image data regardless of whether the image data processing method delivered from the image reading device 10 to the search device 20 is “image data storage” or “image data search”. Although the case where the OCR processing and proper noun extraction technology are applied has been shown, the layout analysis, OCR processing and proper noun extraction technology are performed only in the case of “image data search” and not in the case of “image data storage”. It may be configured. That is, the control unit 21 only stores the image data delivered from the image reading device 10 in the nonvolatile storage unit 25b during the “image data storage” process, and when performing the “image data search” process, The image data stored in the nonvolatile storage unit 25b may be subjected to layout analysis and OCR processing to be searched for text data. Here, in the present embodiment, the case where the data to be accumulated in the “image data accumulation” is image data has been described. However, the accumulated data may be text data subjected to layout analysis and OCR processing. Good. In this case, when performing a search process corresponding to “image data search” in the present embodiment, this text data is directly searched.

  Further, in the present embodiment, the case where a document stored in the nonvolatile storage unit 25b is searched based on a person name is shown. However, the basis is not limited to a person name, and morphological analysis and proper noun extraction technology are used. For example, it may be an organization name or a place name.

  Further, in the present embodiment, the case of performing based on a program stored in advance in the nonvolatile storage unit 25b has been shown, but for example, a CD-ROM (Compact Disk-Read Only Memory), a DVD (Digital Versatile Disk), or the like. The software may be installed in a computer-readable recording medium.

1 is a block diagram showing an overall configuration of a document search system according to an embodiment of the present invention. It is a block diagram which shows an example of the hardware constitutions of a search device. It is the figure which illustrated the format of the proper noun dictionary. It is the figure which illustrated the format of the file creator table. It is the figure which showed the flowchart which showed the operation | movement which the control part of a search device performs. It is the figure which illustrated the selection menu which a display part displays. It is the figure which illustrated the document described in Chinese. It is the figure which illustrated the file name input screen which a display part displays. It is the figure which illustrated the document described in English. It is the figure which illustrated the format of the proper noun dictionary. It is the figure which illustrated the language selection menu which a display part displays.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Document search system, 10 ... Image reading device, 12 ... Communication line, 20 ... Search device, 21 ... Control part, 22 ... Communication IF part, 23 ... Display 24, operation unit, 25 ... storage unit, 25a ... volatile storage unit, 25b ... non-volatile storage unit, 26 ... bus

Claims (1)

First storage means for storing a Chinese name of a person name and an English notation of the person name in association with each other;
First extraction means for extracting a person's name from image data representing a manuscript written in Chinese;
Conversion means for converting a person name extracted by the first extraction means into a person name written in English based on the stored contents of the first storage means;
Accepting means for accepting input of a file name of the manuscript;
Second storage means for storing the file name of the document received by the reception means, the person name extracted by the first extraction means, and the person name converted by the conversion means in association with each other;
Second extraction means for extracting a person's name from image data representing a manuscript written in English;
Search means for searching from the second storage means for the file name of the manuscript associated with the person name extracted by the second extraction means and the person name extracted by the first extraction means; ,
Display means for displaying a file name and a person name searched by the search means;
An information processing apparatus comprising:

Images